parpfish

8 ms·

91.
by parpfish 6mo ago
On one hand, I don’t know why startups make apps. It requires more devs and keeping everything at parity is tough with desktop, iOS, android, mobile web. Seems pragmatic to just simplify and use web. But on the other hand, I’d love to pay y
92.
by parpfish 6mo ago
i think there's also a strong Lynchian influence to the liminal space aesthetic. Industrial spaces, flickering fluoresecnt lights, and a sense of unease that comes from dream-logic and shots that linger to emphasize emptyness.
93.
by parpfish 7mo ago
I can’t deploy a stupid little app at work without something breaking. Im impressed when people can build something so complex that works on the first try.
94.
by parpfish 7mo ago
The output reminds of a really good version of pre-LLM text generation like character lever LSTMs or markov chains. It seems to have syntax down to make superficially good text, but the semantics just aren’t there
95.
by parpfish 7mo ago
after testing, i'm pretty sure that either a) i dont understand Victorian speech very well or b) a model with 340million parameters doesn't generate particularly coherent speech
96.
by parpfish 7mo ago
I don’t want to encourage people touch my screen and smudge things up.
97.
by parpfish 7mo ago
I always thought a great compromise would be games that gave you an overhead “gods eye” third person perspective. People seem to be obsessed with making VR games first person, but that’s where the movement problems come in. The game Moss di
98.
by parpfish 7mo ago
i'd love to have a sort of chatroulette/omegele app for my cats to "socialize" with other cats while I'm away, but it's the internet so I know that it'd get filled with trolls.
99.
by parpfish 7mo ago
The original author said he built this on the weekend, so my assumption is that this was something engineers had advocated for before but were shut down because management wanted them elsewhere. The use of ai agents allowed them to shrink t
100.
by parpfish 7mo ago
I feel so guilty about how often I begged my parents for new magic, Star Wars ccg, or battletech booster packs.
101.
by parpfish 7mo ago
I thought the “daily fantasy” loophole was pretty silly too.
102.
by parpfish 7mo ago
I think we got “lucky” for a while when prediction markets were a small niche interest and they provided a weird novel way of making predictions. But once they get big enough to attract the attention of people that directly affect the outco
103.
by parpfish 7mo ago
When it comes down to it, I’m not sure how you differentiate an “addictive” product from a well-made product that I choose to keep using. When people say that Tetris and Civilization are “addictive” they aren’t implying anything malicious a
104.
by parpfish 7mo ago
A/B testing is one way to make things “addictive” but you can also make addictive products without it. A really good designer could make a highly engaging app or an editor can write clickbait headlines all with without testing.
105.
by parpfish 7mo ago
And you need to find a way to market it that prevents people from thinking “cool, I bet I can get Claude to whip up something similar real quick”
106.
by parpfish 7mo ago
Those names are horrifiq
107.
by parpfish 7mo ago
Something I think about frequently is that 20 years ago, there weren’t machines that could do visual object recognition/categorization and we didn’t really have a clue how humans did it either. We knew that neuron built fancier and fan
108.
by parpfish 7mo ago
As to ye philosophy of “why” the CLT gives you normals, my hunch is that it’s because there’s some connection between: a) the CLT requires samples drawn from a distribution with finite mean and variance and b) the Gaussian is the maximum en
109.
Ask HN: If AI makes your devs super productive, why layoff?
6 points by parpfish 7mo ago|3 comments
110.
by parpfish 7mo ago
If you worked in an office and your boss asked for 100 copies of their memo, they want you to use the copy machine. If they saw you typing it out 100 times they’d tell you that you’re wasting time. It does t matter that you like to type and
111.
by parpfish 7mo ago
I feel like I still get to solve the puzzles I like because I like the higher level architecture/design parts. I just don’t have to type as much because I can provide a stubbed out solution and tell it to fill I the rest.
112.
by parpfish 7mo ago
For debugging, it’s also great trawling through logs and stack traces. Makes a late night oncall page way easier when the bot will tell you exactly what broke
113.
by parpfish 7mo ago
I shouldnt be surprised that my mom is obsessed with her smartphone. As a kid, I remember her talking with friends on the landline phone for what seemed like HOURS
114.
by parpfish 7mo ago
old folks and children both face the same problem with the internet— their initial exposure is to the current internet that has been ab tested into a hyper-addictive hellscape and they are cognitively unprepared. Jumping straight into the
115.
by parpfish 7mo ago
building recommendation systems for classical music has a simple data problem- most recommendation systems (for spotify and others) are based on simple user listening histories that look at "people that listened to X also listened to Y
116.
by parpfish 7mo ago
> they last forever Until my cat finds them. Any serious set of wired headphones better have replaceable wires because apparently they are delicious
117.
by parpfish 7mo ago
note that the teller's job duties shifted as well. with ATMs, they wouldn't hand count money for withdrawals and deposits as much. they'd be doing more interesting and challenging things. same thing will happen with AI automa
118.
by parpfish 7mo ago
i once had this "make a PB&J" as part of a written take-home interview. i knew the schtick -- no matter how precise and complete you are, there is always the possibility for another little gotcha. and that makes it absolute
119.
by parpfish 7mo ago
I think this distinction also gets at some issue with things like privacy and facial recognition. There’s the old approach of hanging a wanted poster and asking people to “call us if you see this guy”. Then there’s the new approach matching
120.
by parpfish 7mo ago
One other way to keep things true to the “20 min max” is to have a clear objective/scoring rubric. Nothing open ended (data science jobs LOVE handing out open ended data analyses). I need to know that it’s okay to stop and that anythin

More ›